A boy once asked Abraham Lincoln how long a man’s legs should be, and Abe replied, “Long enough to reach the ground.”
This past week, mortgage rates just set a new all-time low, about 3.25 percent (“about” because of the always indefinite mixtures of fees at a given rate found by survey). The prior low was not far above, but back in 2012. Before this Great Recession cycle, the prior low was 4.00 percent — the first rate pegged by the VA in 1944.
Rates have risen in the last week — not much — and it’s reasonable to ask if we’ll see a new record, or a series, and how far down. “When” is impossible even to guess, but why we may see new records and how far down are worth studying.
Keeping an eye on the future
First of all, watch the U.S. 10-year T-note (UST-10), not mortgages. The 10-year is the driver, mortgages always following.
The linkage is not exact, as foreign money favors Treasurys, and the mortgage market is often temporarily congested by waves of refinancing locks.
But over time, the relationship of mortgages to UST-10s is solid. Two forces establish spreads relative to Treasurys: credit risk and market risk.
We can isolate mortgage credit risk by comparing the rates on old Fannie straight debt (simple IOUs with fixed 10-year maturity) to UST-10s — and that, spread over a very long time, has been close to 0.30 percent.
However, over that same time, the overall spread UST-10s to retail 30-fixed mortgages has been about 1.80 percent. What’s the other 1.50 percent for, the premium over-credit risk?
A little more than 0.25 percent is the cost to service the loan. The investor-owner of a loan must cough up some yield to hire somebody to collect the payments. And there’s a like amount to securitize into MBSs (mortgage-backed securities).
Okay, that still leaves a fat 1.00 percent spread on the table.
That’s the privilege of prepayment at will. Unlimited prepayment presents to investors as a lose-lose deal. If rates rise, I get killed.
But I can hedge that, maintaining a short position against some of my long holdings.
But, if rates fall, then these rotten consumers refinance and I’m left with no loans and just upside-down hedges.
So: You can have open prepayment, but it will cost you 1.00 percent.
How low can they go?
So, let’s ask the right question. How low can UST-10s go? No matter where they go, they will drag mortgages behind them, up or down.
On July 8, UST-10s set their all-time low, too, at 1.36 percent.
How low, oh Lord, how low? In that same week, yields on German 10-year “Bunds” reached their all-time low, too, at minus-0.20 percent. (A pattern developing: Bunds pulling USTs the way USTs pull mortgages.)
If conditions in the U.S. one day resemble those in Europe — a deflationary depression afflicting more than half of the continent, our central bank in frantic easing like the European Central Bank — there is no reason that UST-10s could not fall to or below zero.
If our economy — when our economy — slides again toward recession, and if we have not had a recovery strong enough to pull up our whole rate structure, and rates fall from here….UST-10s will begin to slide, and mortgages will follow 1.80 percent behind.
If UST-10s fall to zero someday, then U.S. mortgages could trade below 2.00 percent.
Be very careful what we wish for.
Lou Barnes is a mortgage broker based in Boulder, Colorado. He can be reached at lbarnes@pmglending.com.
